- Email: [email protected]
A dosimetry system for boron neutron capture therapy based on the dual counter microdosimetric technique
173s
Bull CancerlRadiother 1996;83 (Suppl I): 173s-175s © Elsevier, Paris
Physics and dosimetry
A dosimetry system for boron neutron capture therapy based on the dual counter microdosimetric technique C Kota, RL Maughan Gershenson Radiation Oncology Center, Harper Hospital and Wayne State University, 3990 John R, Detroit, M148201, USA
Summary - A dosimetric technique based on proportional counter microdosimetry has been developed for the dosimetry of mixed radiation fields employed in boron neutron capture therapy (BNCT). The technique has been successfully used to measure the gamma, fast neutron and the boron dose rates in the mixed radiation field from a 252Cf source. The measured fast neutron and boron dose rates are in good agreement with the calculations and the measurements reported by other researchers. A systematic discrepancy is observed in the gamma dose rate measurements, with the present measurements being approximately 35% lower than those reported elsewhere. BNCT dosimetry I microdosimetry I dual counter technique Resume - Dosimetrie dans Ie traitement par capture de neutrons par Ie bore: technique microdosimetrique des doubles compteurs. Une technique de dosimetrie basee sur des compteurs proportionnels microdosimetriques a ete developpee pour la dosimetrie desfaisceaux d'irradiation mixtes utilises dans Ie traitement par capture des neutrons par Ie bore (BNCT). Cette technique a ete utilisee avec succes pour mesurer les debits de doses de y, de neutrons rapides et des particules secondaires a la capture dans un faisceau emis par une source de californium 252. Les debits de doses mesures pour les neutrons rapides et les particules secondaires a la capture sont en accord avec les resultats calcules et les mesures realisees par d'autres equipes. Une difference systematique est observee dans les mesures du debit de dose r avec des valeurs de 35 % inferieures a celles rapportees par les autres auteurs. BNCT dosimetrie I microdosimetrie I technique des doubles compteurs
INTRODUCTION In recent years there has been renewed interest in boron neutron capture therapy (BNCT), spurred by the development of tumor-specific drugs and by the development of reactor- and accelerator-based epithermal neutron beams [1, 3]. The boron neutron capture reaction might also be useful in providing a local tumor dose enhancement in fast neutron therapy or 252 californium (252Cf) brachytherapy so as to cause a clinically useful shift in the tumor control probability curve [2]. In all of the above situations, the radiation field of interest consists of gamma rays, fast neutrons, epithermal neutrons and thermal neutrons. A dosimetric description of these radiation fields involves a
measurement of the different dose components at various spatial positions in suitable phantoms. A dosimetry technique using tissue equivalent (TE) low pressure proportional counters has been developed for the direct measurement of the absorbed dose from the different components of the radiation field. The technique has been used for the measurement of the dose contributions from the different components of the mixed radiation field produced by 252Cf. The measurements have been compared with Monte Carlo calculations and measurements made with other experimental dosimetry techniques. The basic uncertainties associated with this method in the measurement of the absorbed dose have been briefly discussed.
174s
C Kota, RL Maughan
MATERIALS AND METHODS The dosimetry system is based on the measurement of the microdosimetric single event spectra at the point of interest in the radiation field with two identical 1/2" LET counters. The counters have 8-mm thick tissue equivalent plastic (TEP) A-150 walls (manufactured by Far West Technologies, Goleta, CA) and an internal diameter of 1.27 cm. The A-150 TEP wall of one of the counters is uniformly loaded with 50 Jlglg of boron (lOB) to simulate a tissue containing a uniform distribution of 50 p.g/g of lOB. The two counters are simultaneously filled with a propane-based TE gas (C0 2: 40%, N2: 5%, C 3Hg : 55%) to a pressure of 2.2 kPa to simulate a unit density tissue volume of 0.5 Jlm and are operated at a positive voltage of 550 V. The counters are energy-calibrated by positioning the proton edge in the measured spectra at a lineal energy value of 150.0 keV/Jlm. The two counters were irradiated simultaneously under identical conditions and the lineal energy spectra were acquired using a PC-based data acquisition system described elsewhere [2]. The 252Cf source was held in the middle of a l-cm thick Lucite-walled water phantom of dimensions 30 x 30 x 60 cm by a Lucite holder. The counters were positioned symmetrically about the source in its transverse plane using a Lucite jig with a reproducibility of - 3 mm. The counter with no lOB in its wall (TE counter) was used to measure the complete spectrum of the absorbed dose from the 252Cf source over the lineal energy range of -1.5 to -1100.0 keV/Jlm. The measurement of the spectra by the TE counter was limited by the system noise to -1.5 keVI p.m. These spectra were extrapolated to lower lineal energy values by fitting a gamma spectrum estimated from a predominantly gamma spectrum measured at a distance of 15 cm from the source to a lineal energy value of -0.6 keV/JlM. To establish the accuracy of the gamma dose measurement, the dose absorbed by the TE counter placed at a distance of 25 cm from a calibrated 137CS source was measured and compared to the expected dose. The neutron dose was obtained by subtracting the dose from the fitted gamma spectrum from the total dose measured with the TE counter, The lineal energy spectrum of the counter with lOB in its wall (lOB counter) was measured over the range of -10.0 to -1100.0 keV/Jlm. To obtain the dose deposited by alpha particles and the recoil lithium ions resulting from the boron neutron capture (BNC) reaction (henceforth referred to as the BNC dose), the lOB counter spectrum was matched to the TE counter spectrum in the lineal energy interval of -15.0 to -35.0 keV/p.m where no contribution from the BNC dose is expected. Disagreements between the doses measured by the two counters in this lineal energy interval were attributed to uncertainties in counter positioning and the dose measured by the
4.0E5 TE counter spectrum 108 counter spectrum gamma dose spectrum
3.0E5 Q)
III 0
c
>.
proton edge
2.0E5
"'C
>.
1.0E5
10-2
10-'
100
101
102
103
lineal energy y (keV f-llll-1) Fig 1. Microdosimetric single event spectra measured at a distance of 4 cm from the center of a 252Cf source along its transverse axis. - - : TE counter spectrum; - - - -: lOB counter spectrum; - - - - : gamma spectrum. The position of the proton edge at ISO keV/,um is used to self-calibrate the TE counter spectrum.
lOB counter was normalized to that measured by the TE counter.
RESULTS Figure 1 shows the microdosimetric spectra measured with the two counters and the gamma spectrum used to separate the gamma dose. The TE counter spectrum is marked by a distinct proton edge at 150 keY/pm which was used for the energy loss calibration of the counter. The spectrum measured by the lOB counter shows the dose deposited by the alpha particles and the recoil lithium ions produced in the lOB thermal neutron capture reaction. The lineal energy spectrum of the BNC dose has a peak at a lineal energy of about 320 keV/J.lm and extends to lower lineal energies of -35 keV/J.lm. The overlap of the dose deposited by these reaction products with that of the protons at the proton edge shifts the edge to a higher lineal energy value, as seen in figure 1. Figure 2 shows the gamma, neutron and boron dose rates in tissue, normalized to a source strength of 1 J.lglhour, as a function of the distance from the source. The horizontal error bars represent the 3 mm positional error associated with the measurements. The vertical error bars on the gamma and the neutron dose rate measurements represent a 7% uncertainty arising from the calibration uncertainties (5%) and the gamma dose separation procedure (5%). The vertical error bars on the BNC dose rate measurements represent a 7% uncertainty arising from calibration uncertainties (5%) and the uncertainty associated with the use of a constant W value in the calibration procedure. The measured gamma
175s
A dosimetry system for BNCf 1 ()O
r:-------------------, netrtron rto.e
'08 do..
gammados. WIerzbicki., aL(neulron dose) YlIlCh ., al. (neutron do..) WIerzbicki ., aL(' OS dose)
2
10'1
2
10-2
..... 4-. 2
3
-4
5
6
7
8
9
10
11
distance from source (em)
Fig 2. Gamma. neutron and lOB dose rates in tissue measured by the dual counter microdosimetric technique. The horizontal error bars represent 3 mm positioning uncertainty while the vertical error bars represent 7 % random uncertainty in the measurements.
obtained from the present measurements show a substantial disagreement with those reported by Yanch et al and Wierzbicki et aI, and are systematically lower by about 35%. Since the gamma dose component measurement was verified by measuring the dose from a calibrated 137CS source under identical TE counter operating conditions, it is unlikely that the discrepancy is the result of a systematic error in the measurements. The cause of the discrepancy is being investigated. The calculation of the absorbed dose in the tissue of interest from the microdosimetric measurements involves a correction for differences in kerma factors and mass attenuation coefficients between A-I50 TEP and the tissue of interest, which necessitates an a priori knowledge of the energy spectrum of the radiation field. In principle, this limitation can be overcome by the use of a wall material better matched to the tissue of interest in terms of atomic composition. The dosimetry system in its present form is also limited by count rate requirements to use in radiation fields with low dose rates, eg, accelerator produced epithermal beams. However, the technique can be easily extended for use in fields with higher dose rates by using paired counters with smaller physical dimensions.
dose rates were corrected for the differences in the W values between the calibration protons and the secondary electrons by using a correction factor of 0.95. The measured neutron dose was converted to that in tissue using a spectrum weighted kerma factor ratio of 0.983. The measured boron dose rates were corrected for an estimated 5% under-response of the counter associated with the absence of lOB in the TE gas. The neutron and boron dose rates reported by other researchers are shown in comparison with the present measurements.
We would like to thank AT Porter for his support and encouragement of this work.
DISCUSSION AND CONCLUSION
REFERENCES
The present measurements of the neutron dose rates ace in good agreement with the Monte Carlo calculations reported by Yanch et al [5] and the paired TE/Mg counter measurements reported by Wierzbicki et al [4]. The observed discrepancies are within the limits of experimental error. The boron dose rates are also in fair agreement with the dose rate values reported by Wierzbicki et al for water. The thermal neutron fluence in A-I50 TEP can be expected to be different from that in water due to differences in the atomic composition, which might explain the observed differences in the boron dose rates. The gamma dose rates
ACKNOWLEDGMENT
2
3 4 5
Allen Bl, Moore DE, Harrington BV. Progress in Neutron Capture Therapy for Cancer. New York: Plenum Press, 1992 Laramore GE, Wootton P, Livesey lC et aI. Boron neutron capture therapy: a mechanism for achieving a concomitant tumor boost in fast neutron radiotherapy. Int J Radial Oncol Bioi Phys 1994;28:1125--42 Soloway AH, Banh RF, Carpenter DE. Advances in NeuIron Capture Therapy. New York: Plenum Press, 1993 Wierzbicki lG, Maruyama Y, Porter AT. Measurement of augmentation of 252Cf implant by lOB and 157Gd neutron capture. Med Phys 1994;21:787-90 Yanch lC, Zamenhof RG. Dosimetry of 252Cf implants for neutron radiotherapy with and without augmentation by boron neutron capture therapy. Radiat Res 1992; 131:249-56
Bull CancerlRadiother 1996;83 (Suppl I): 173s-175s © Elsevier, Paris
Physics and dosimetry
A dosimetry system for boron neutron capture therapy based on the dual counter microdosimetric technique C Kota, RL Maughan Gershenson Radiation Oncology Center, Harper Hospital and Wayne State University, 3990 John R, Detroit, M148201, USA
Summary - A dosimetric technique based on proportional counter microdosimetry has been developed for the dosimetry of mixed radiation fields employed in boron neutron capture therapy (BNCT). The technique has been successfully used to measure the gamma, fast neutron and the boron dose rates in the mixed radiation field from a 252Cf source. The measured fast neutron and boron dose rates are in good agreement with the calculations and the measurements reported by other researchers. A systematic discrepancy is observed in the gamma dose rate measurements, with the present measurements being approximately 35% lower than those reported elsewhere. BNCT dosimetry I microdosimetry I dual counter technique Resume - Dosimetrie dans Ie traitement par capture de neutrons par Ie bore: technique microdosimetrique des doubles compteurs. Une technique de dosimetrie basee sur des compteurs proportionnels microdosimetriques a ete developpee pour la dosimetrie desfaisceaux d'irradiation mixtes utilises dans Ie traitement par capture des neutrons par Ie bore (BNCT). Cette technique a ete utilisee avec succes pour mesurer les debits de doses de y, de neutrons rapides et des particules secondaires a la capture dans un faisceau emis par une source de californium 252. Les debits de doses mesures pour les neutrons rapides et les particules secondaires a la capture sont en accord avec les resultats calcules et les mesures realisees par d'autres equipes. Une difference systematique est observee dans les mesures du debit de dose r avec des valeurs de 35 % inferieures a celles rapportees par les autres auteurs. BNCT dosimetrie I microdosimetrie I technique des doubles compteurs
INTRODUCTION In recent years there has been renewed interest in boron neutron capture therapy (BNCT), spurred by the development of tumor-specific drugs and by the development of reactor- and accelerator-based epithermal neutron beams [1, 3]. The boron neutron capture reaction might also be useful in providing a local tumor dose enhancement in fast neutron therapy or 252 californium (252Cf) brachytherapy so as to cause a clinically useful shift in the tumor control probability curve [2]. In all of the above situations, the radiation field of interest consists of gamma rays, fast neutrons, epithermal neutrons and thermal neutrons. A dosimetric description of these radiation fields involves a
measurement of the different dose components at various spatial positions in suitable phantoms. A dosimetry technique using tissue equivalent (TE) low pressure proportional counters has been developed for the direct measurement of the absorbed dose from the different components of the radiation field. The technique has been used for the measurement of the dose contributions from the different components of the mixed radiation field produced by 252Cf. The measurements have been compared with Monte Carlo calculations and measurements made with other experimental dosimetry techniques. The basic uncertainties associated with this method in the measurement of the absorbed dose have been briefly discussed.
174s
C Kota, RL Maughan
MATERIALS AND METHODS The dosimetry system is based on the measurement of the microdosimetric single event spectra at the point of interest in the radiation field with two identical 1/2" LET counters. The counters have 8-mm thick tissue equivalent plastic (TEP) A-150 walls (manufactured by Far West Technologies, Goleta, CA) and an internal diameter of 1.27 cm. The A-150 TEP wall of one of the counters is uniformly loaded with 50 Jlglg of boron (lOB) to simulate a tissue containing a uniform distribution of 50 p.g/g of lOB. The two counters are simultaneously filled with a propane-based TE gas (C0 2: 40%, N2: 5%, C 3Hg : 55%) to a pressure of 2.2 kPa to simulate a unit density tissue volume of 0.5 Jlm and are operated at a positive voltage of 550 V. The counters are energy-calibrated by positioning the proton edge in the measured spectra at a lineal energy value of 150.0 keV/Jlm. The two counters were irradiated simultaneously under identical conditions and the lineal energy spectra were acquired using a PC-based data acquisition system described elsewhere [2]. The 252Cf source was held in the middle of a l-cm thick Lucite-walled water phantom of dimensions 30 x 30 x 60 cm by a Lucite holder. The counters were positioned symmetrically about the source in its transverse plane using a Lucite jig with a reproducibility of - 3 mm. The counter with no lOB in its wall (TE counter) was used to measure the complete spectrum of the absorbed dose from the 252Cf source over the lineal energy range of -1.5 to -1100.0 keV/Jlm. The measurement of the spectra by the TE counter was limited by the system noise to -1.5 keVI p.m. These spectra were extrapolated to lower lineal energy values by fitting a gamma spectrum estimated from a predominantly gamma spectrum measured at a distance of 15 cm from the source to a lineal energy value of -0.6 keV/JlM. To establish the accuracy of the gamma dose measurement, the dose absorbed by the TE counter placed at a distance of 25 cm from a calibrated 137CS source was measured and compared to the expected dose. The neutron dose was obtained by subtracting the dose from the fitted gamma spectrum from the total dose measured with the TE counter, The lineal energy spectrum of the counter with lOB in its wall (lOB counter) was measured over the range of -10.0 to -1100.0 keV/Jlm. To obtain the dose deposited by alpha particles and the recoil lithium ions resulting from the boron neutron capture (BNC) reaction (henceforth referred to as the BNC dose), the lOB counter spectrum was matched to the TE counter spectrum in the lineal energy interval of -15.0 to -35.0 keV/p.m where no contribution from the BNC dose is expected. Disagreements between the doses measured by the two counters in this lineal energy interval were attributed to uncertainties in counter positioning and the dose measured by the
4.0E5 TE counter spectrum 108 counter spectrum gamma dose spectrum
3.0E5 Q)
III 0
c
>.
proton edge
2.0E5
"'C
>.
1.0E5
10-2
10-'
100
101
102
103
lineal energy y (keV f-llll-1) Fig 1. Microdosimetric single event spectra measured at a distance of 4 cm from the center of a 252Cf source along its transverse axis. - - : TE counter spectrum; - - - -: lOB counter spectrum; - - - - : gamma spectrum. The position of the proton edge at ISO keV/,um is used to self-calibrate the TE counter spectrum.
lOB counter was normalized to that measured by the TE counter.
RESULTS Figure 1 shows the microdosimetric spectra measured with the two counters and the gamma spectrum used to separate the gamma dose. The TE counter spectrum is marked by a distinct proton edge at 150 keY/pm which was used for the energy loss calibration of the counter. The spectrum measured by the lOB counter shows the dose deposited by the alpha particles and the recoil lithium ions produced in the lOB thermal neutron capture reaction. The lineal energy spectrum of the BNC dose has a peak at a lineal energy of about 320 keV/J.lm and extends to lower lineal energies of -35 keV/J.lm. The overlap of the dose deposited by these reaction products with that of the protons at the proton edge shifts the edge to a higher lineal energy value, as seen in figure 1. Figure 2 shows the gamma, neutron and boron dose rates in tissue, normalized to a source strength of 1 J.lglhour, as a function of the distance from the source. The horizontal error bars represent the 3 mm positional error associated with the measurements. The vertical error bars on the gamma and the neutron dose rate measurements represent a 7% uncertainty arising from the calibration uncertainties (5%) and the gamma dose separation procedure (5%). The vertical error bars on the BNC dose rate measurements represent a 7% uncertainty arising from calibration uncertainties (5%) and the uncertainty associated with the use of a constant W value in the calibration procedure. The measured gamma
175s
A dosimetry system for BNCf 1 ()O
r:-------------------, netrtron rto.e
'08 do..
gammados. WIerzbicki., aL(neulron dose) YlIlCh ., al. (neutron do..) WIerzbicki ., aL(' OS dose)
2
10'1
2
10-2
..... 4-. 2
3
-4
5
6
7
8
9
10
11
distance from source (em)
Fig 2. Gamma. neutron and lOB dose rates in tissue measured by the dual counter microdosimetric technique. The horizontal error bars represent 3 mm positioning uncertainty while the vertical error bars represent 7 % random uncertainty in the measurements.
obtained from the present measurements show a substantial disagreement with those reported by Yanch et al and Wierzbicki et aI, and are systematically lower by about 35%. Since the gamma dose component measurement was verified by measuring the dose from a calibrated 137CS source under identical TE counter operating conditions, it is unlikely that the discrepancy is the result of a systematic error in the measurements. The cause of the discrepancy is being investigated. The calculation of the absorbed dose in the tissue of interest from the microdosimetric measurements involves a correction for differences in kerma factors and mass attenuation coefficients between A-I50 TEP and the tissue of interest, which necessitates an a priori knowledge of the energy spectrum of the radiation field. In principle, this limitation can be overcome by the use of a wall material better matched to the tissue of interest in terms of atomic composition. The dosimetry system in its present form is also limited by count rate requirements to use in radiation fields with low dose rates, eg, accelerator produced epithermal beams. However, the technique can be easily extended for use in fields with higher dose rates by using paired counters with smaller physical dimensions.
dose rates were corrected for the differences in the W values between the calibration protons and the secondary electrons by using a correction factor of 0.95. The measured neutron dose was converted to that in tissue using a spectrum weighted kerma factor ratio of 0.983. The measured boron dose rates were corrected for an estimated 5% under-response of the counter associated with the absence of lOB in the TE gas. The neutron and boron dose rates reported by other researchers are shown in comparison with the present measurements.
We would like to thank AT Porter for his support and encouragement of this work.
DISCUSSION AND CONCLUSION
REFERENCES
The present measurements of the neutron dose rates ace in good agreement with the Monte Carlo calculations reported by Yanch et al [5] and the paired TE/Mg counter measurements reported by Wierzbicki et al [4]. The observed discrepancies are within the limits of experimental error. The boron dose rates are also in fair agreement with the dose rate values reported by Wierzbicki et al for water. The thermal neutron fluence in A-I50 TEP can be expected to be different from that in water due to differences in the atomic composition, which might explain the observed differences in the boron dose rates. The gamma dose rates
ACKNOWLEDGMENT
2
3 4 5
Allen Bl, Moore DE, Harrington BV. Progress in Neutron Capture Therapy for Cancer. New York: Plenum Press, 1992 Laramore GE, Wootton P, Livesey lC et aI. Boron neutron capture therapy: a mechanism for achieving a concomitant tumor boost in fast neutron radiotherapy. Int J Radial Oncol Bioi Phys 1994;28:1125--42 Soloway AH, Banh RF, Carpenter DE. Advances in NeuIron Capture Therapy. New York: Plenum Press, 1993 Wierzbicki lG, Maruyama Y, Porter AT. Measurement of augmentation of 252Cf implant by lOB and 157Gd neutron capture. Med Phys 1994;21:787-90 Yanch lC, Zamenhof RG. Dosimetry of 252Cf implants for neutron radiotherapy with and without augmentation by boron neutron capture therapy. Radiat Res 1992; 131:249-56